The PANHIS Study: Sexual Health in Spanish-Speaking Adults - A Cross-Sectional Online Survey

Overview

This cross-sectional observational study uses an online survey to describe the epidemiology of sexual health, sexual behaviour and sexual function in Spanish-speaking adults, and to estimate the prevalence and phenomenology of expanded (altered) states of consciousness associated with orgasm. The study has two prespecified aims, analysed and reported as two separate manuscripts: 1. sexual health and behaviour epidemiology and associated factors; and 2. expanded states of consciousness associated with orgasm. Recruitment is via social media with complementary channels. Sexual function is assessed with a brief study-developed instrument administered identically to all participants, enabling direct between-sex comparison. Survey weights are applied. Reporting follows STROBE and CHERRIES.

Background and rationale Sexual health is recognised by the World Health Organization (2006) as a state of physical, emotional, mental, and social well-being in relation to sexuality, beyond the absence of disease. Large-scale epidemiological surveys such as the National Survey of Sexual Attitudes and Lifestyles (Natsal) in the United Kingdom and the National Health and Social Life Survey (NHSLS) in the United States have provided fundamental evidence informing public health policy. No equivalent study currently exists for the Spanish-speaking world covering sexual health cross-nationally. In addition, an emerging body of neuroscience literature has documented activation of brain networks compatible with trance and mystical experiences during orgasm (Komisaruk et al., 2004; Safron, 2016). The prevalence, associated factors, and subjective impact of such expanded states of consciousness in the general Spanish-speaking population have not been systematically studied. Study design This is a multi-country, cross-sectional, observational, descriptive study. Data are collected through a single online questionnaire administered via a custom web application. Submissions are transmitted to a Google Form, which writes them automatically to a linked, EU-hosted spreadsheet; this architecture was chosen because it withstands high-concurrency traffic spikes following social media posts without silent loss of responses, which a direct-write backend would not reliably withstand. Data are processed in compliance with the GDPR. Participation is anonymous for participants who do not provide an email address, and pseudonymised for those who choose to provide one, which is used only for assistance with the study incentive and, subject to a separate and independent consent, for recontact regarding future studies. Population and recruitment The target population comprises adults (≥18 years) who are Spanish-speaking and have internet access. Sampling is non-probabilistic, by convenience, primarily through the Instagram account @mamacasquet (\~2.5 million followers) with a video introduction and a direct link to the survey. Complementary recruitment is conducted via Facebook, X, allied social media accounts, and professional sexology associations to diversify the sample by age and country. Eligibility for participation is detailed under Eligibility Criteria. In addition to those criteria, the following data-quality rules are applied at analysis: completion rate below 70% of the questionnaire, which is enforced by the survey platform itself and prevents submission; duplicate responses (detected by exact email match where provided, or by response pattern and technical fingerprint); and implausible response patterns (for example, age of sexual debut greater than current age, systematic straight-lining on reverse-coded items, or completion time below the first percentile of the pilot distribution). Except for the 70% threshold, records failing these checks are retained in the dataset and flagged for sensitivity analysis rather than excluded outright. Sample size justification To estimate prevalences in the general population, assuming the most conservative scenario (p = 0.5), 95% confidence level, and ±1.4 percentage point margin of error, the minimum sample size required is approximately 5,000 complete responses. For country-stratified analyses (assuming 8-10 countries with sufficient representation and prevalences around 30%), 500 responses per stratum yield precision of ±4 percentage points. For multivariable logistic regression models with 15-20 predictors, applying the rule of thumb of 10-20 events per variable (Peduzzi et al., 1996), the minimum sample size is comfortably met assuming event prevalences ≥ 5%. Given the audience of the principal collaborator (\~2.5 million followers) and expected response rates of 0.5-1%, the anticipated final sample size is 15,000 to 30,000 complete responses. Measures Sexual function and satisfaction are assessed with a brief instrument developed by the research team (the PANHIS Sexual Function Index, PSFI): 11 items administered identically to all participants, with a common 4-week recall period and a common 0-4 response scale, covering frequency of sexual activity, desire, subjective arousal, genital physical response, orgasm, pain during receptive penetration, sexual satisfaction, associated distress, global self-rated sexual function, sexual self-esteem, and communication with partner(s). Rationale for a study-developed instrument. The protocol originally approved by the Research Ethics Committee specified the Female Sexual Function Index short version (FSFI-6) for participants with female sexual response and the International Index of Erectile Function (IIEF-5) for participants with male sexual response. Licence applications for academic use of both instruments, submitted through Mapi Research Trust, received no response and no timeframe could be established; this change was notified to the Committee before data collection began. Beyond availability, administering different instruments to men and women confounds between-sex differences with differences between the instruments themselves, since the FSFI-6 and IIEF-5 differ in recall period, item count, domain structure and cut-off calibration. An identical item set with an identical recall period and response scale permits direct and interpretable between-sex comparison. In addition, the PSFI treats absence of sexual activity during the recall period as not applicable rather than as a zero score, avoiding the overestimation artefact documented for instruments that assign null scores to sexual inactivity. The phenomenological module on expanded states of consciousness during orgasm is conceptually informed by domains described in the Mystical Experience Questionnaire (MEQ-30; MacLean et al., 2012) and in the broader literature on non-ordinary states of consciousness, but does not constitute a formal application of the MEQ-30 or of any other standardised instrument; these items are of original wording and are prespecified as exploratory. The remaining items (sociodemographics, general health, sexual behaviour, education, psychosocial factors) are ad hoc and were developed by the research team. Data quality is additionally supported by two embedded attention/validity items, distributed at separated points in the questionnaire, and by age verification enforced at three points in the questionnaire flow. Participants failing the attention items are retained in the dataset and flagged for sensitivity analysis rather than excluded. Measurement properties of the study-developed index will also be examined in this sample, including internal consistency, dimensionality (exploratory factor analysis in a randomly selected half of the sample and confirmatory factor analysis in the other half, with a fixed random seed recorded in the analysis code), convergent validity against a global self-rated anchor item, floor and ceiling effects, and measurement invariance across sex, age group and country where sample size permits. These analyses provide preliminary psychometric evidence and do not constitute a complete validation study: the cross-sectional design precludes assessment of test-retest reliability and responsiveness, and no criterion measure is administered. Statistical analysis plan Descriptive analyses (frequencies, means, medians, standard deviations). Bivariate analyses (chi-square, t-tests, ANOVA, Spearman correlations). Multivariable logistic regression for factors associated with self-reported sexual difficulty with associated distress and with expanded states of consciousness, adjusting for prespecified confounders (age, sex recorded at birth, sexual identity, sexual orientation, country, education level, relationship status, physical and mental health comorbidities, alcohol and substance use). Post-stratification weighting by age and sex according to national census data. The false discovery rate is controlled using the Benjamini-Hochberg procedure within each axis's hypothesis family. Sensitivity analyses assess the impact of selection bias, weighting, attention-item failure, and completion-time outliers. Quantitative variables are analysed both as continuous and as categorised according to prespecified cut-offs. Missing data: complete-case analysis as primary; multiple imputation by chained equations (MICE, 20 imputations, Rubin's rules) as sensitivity analysis if missingness exceeds 5% on key variables, except for "not applicable" responses (for example, absence of sexual activity), which are never imputed. Software: R (v4.x) and/or SPSS (v28+). Reporting and data sharing The study will be reported in accordance with the STROBE guidelines (von Elm et al., 2007) for cross-sectional observational studies and the CHERRIES guidelines for internet-based surveys. After publication of the main manuscript, anonymised data (without the email column) and analysis code will be deposited in a public repository (Zenodo or equivalent) under an open licence, subject to data protection requirements.